Table 2.
95% Coverage Probabilities for a Sample Size of 50 With Type 3 Distribution.
| ρ |
.30 |
.56 |
Unstructured |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| IRC | k | 6 | 12 | 18 | 24 | 6 | 12 | 18 | 24 | 6 | 12 | 18 | 24 |
| 2 | DTLG | .960 | .960 | .972 | .972 | .965 | .968 | .954 | .964 | .955 | .967 | .955 | .968 |
| PRLG | .935 | .944 | .941 | .945 | .935 | .954 | .944 | .926 | .933 | .952 | .929 | .945 | |
| PB | .956 | .964 | .958 | .961 | .969 | .956 | .945 | .942 | .958 | .960 | .942 | .953 | |
| BCa | .950 | .964 | .964 | .968 | .960 | .970 | .954 | .948 | .958 | .958 | .952 | .959 | |
| NTB | .959 | .956 | .965 | .967 | .955 | .969 | .960 | .959 | .941 | .962 | .954 | .958 | |
| BTLG | .971 | .974 | .977 | .982 | .970 | .987 | .977 | .975 | .962 | .977 | .972 | .976 | |
| 3 | DTLG | .922 | .939 | .923 | .947 | .944 | .924 | .927 | .939 | .923 | .930 | .927 | .933 |
| PRLG | .936 | .940 | .928 | .949 | .933 | .926 | .932 | .932 | .953 | .923 | .937 | .929 | |
| PB | .943 | .945 | .923 | .942 | .955 | .921 | .927 | .928 | .945 | .933 | .934 | .927 | |
| BCa | .929 | .946 | .930 | .946 | .938 | .934 | .939 | .945 | .929 | .942 | .942 | .936 | |
| NTB | .942 | .957 | .948 | .961 | .943 | .955 | .967 | .958 | .936 | .940 | .952 | .952 | |
| BTLG | .958 | .969 | .952 | .965 | .956 | .969 | .974 | .968 | .956 | .964 | .956 | .960 | |
| 5 | DTLG | .925 | .936 | .920 | .934 | .919 | .930 | .919 | .924 | .937 | .934 | .935 | .943 |
| PRLG | .937 | .940 | .924 | .943 | .916 | .935 | .927 | .915 | .938 | .917 | .935 | .933 | |
| PB | .934 | .941 | .915 | .936 | .928 | .930 | .920 | .922 | .939 | .932 | .928 | .937 | |
| BCa | .934 | .947 | .917 | .944 | .916 | .938 | .919 | .926 | .933 | .926 | .933 | .938 | |
| NTB | .944 | .952 | .944 | .957 | .923 | .951 | .954 | .952 | .937 | .943 | .953 | .959 | |
| BTLG | .953 | .960 | .938 | .957 | .946 | .968 | .966 | .953 | .950 | .951 | .952 | .951 | |
| 7 | DTLG | .930 | .942 | .946 | .942 | .941 | .943 | .934 | .921 | .924 | .934 | .928 | .936 |
| PRLG | .933 | .925 | .945 | .922 | .941 | .940 | .926 | .922 | .913 | .936 | .918 | .927 | |
| PB | .935 | .941 | .933 | .922 | .933 | .936 | .935 | .922 | .927 | .933 | .919 | .924 | |
| BCa | .924 | .943 | .938 | .923 | .926 | 943 | .934 | .932 | .914 | .934 | .919 | .929 | |
| NTB | .929 | .949 | .956 | .953 | .944 | .960 | .952 | .953 | .933 | .954 | .954 | .945 | |
| BTLG | .942 | .954 | .952 | .940 | .955 | .968 | .958 | .953 | .939 | .956 | .943 | .951 | |
Note. For IRC = 2: skewness = 0.41 and kurtosis = −1.83. For IRC = 3, 5, 7: skewness = 0.97 and kurtosis = −.20. DTLG = delta method with logit transformation; PRLG = three-step parceling with logit transformation; PB = percentile bootstrap; BCa = biased-corrected and accelerated bootstrap; NTB = normal theory bootstrap; BTLG = bootstrap SE with logit transformation. Unacceptable coverage is bolded and outside [.925, .975]. All methods based on 2,000 bootstrap samples and 1,000 simulation replications.